检索结果-南通市图书馆

Clustering in the Wireless Channel with a Power Weighted Statistical Mixture Model in Indoor Scenario

在线全文

学校读者我要写书评

暂无评论

China Communications 2019年第7期16卷 83-95页

作者： Yupeng Li Jianhua Zhang Pan Tang Lei Tian Key Lab of Universal Wireless Communications Beijing University of Posts and Telecommunications Beijing 100876 China State Key Lab of Networking and Switching Technology Beijing University of Posts and Telecommunications Beijing 100876 China

Cluster-based channel model is the main stream of fifth generation mobile communications, thus the accuracy of clustering algorithm is important. Traditional Gaussian mixture model (GMM) does not consider the power information which is important for the channel multipath clustering. In this paper, a normalized power weighted GMM (PGMM) is introduced to model the channel multipath components (MPCs). With MPC power as a weighted factor, the PGMM can fit the MPCs in accordance with the cluster-based channel models. Firstly, expectation maximization (EM) algorithm is employed to optimize the PGMM parameters. Then, to further increase the searching ability of EM and choose the optimal number of components without resort to cross-validation, the variational Bayesian (VB) inference is employed. Finally, 28 GHz indoor channel measurement data is used to demonstrate the effectiveness of the PGMM clustering algorithm.

关键词： channel multipath clustering mmWave Gaussian mixture model expectation maximization variational Bayesian inference

Interpolation mechanism of fuzzy control

在线全文

学校读者我要写书评

暂无评论

Science China(Technological Sciences) 1998年第3期41卷 312-320页

作者：李洪兴 Department of Mathematics Beijing Normal University Beijing China

The fuzzy control algorithms used commonly at present are all regarded as some interpolation functions, which is in essence equivalent to discrete response functions to be fitted. This means that fuzzy control method is similar to finite element method in mathematical physics, which is a kind of direct manner or numerical method in control systems.

关键词： fuzzy control fuzzy inference base functions interpolation functions.

Fuzzy-GA PID controller with incomplete derivation and its application to intelligent bionic artificial leg

在线全文

学校读者我要写书评

暂无评论

Journal of Central South University of Technology 2003年第3期10卷 237-243页

作者：谭冠政李安平 School of Information Science and Engineering Central South University Changsha 410083 China

An optimal PID controller with incomplete derivation is proposed based on fuzzy inference and the geneticalgorithm, which is called the fuzzy-GA PID controller with incomplete derivation. It consists of the off-line part andthe on-line part. In the off-line part, by taking the overshoot, rise time, and settling time of system unit step re-sponse as the performance indexes and by using the genetic algorithm, a group of optimal PID parameters K*p , Ti* ,and Tj are obtained, which are used as the initial values for the on-line tuning of PID parameters. In the on-linepart, based on K; , Ti* , and T*d and according to the current system error e and its time derivative, a dedicatedprogram is written, which is used to optimize and adjust the PID parameters on line through a fuzzy inference mech-anism to ensure that the system response has optimal dynamic and steady-state performance. The controller has beenused to control the D. C. motor of the intelligent bionic artificial leg designed by the authors. The result of computersimulation shows that this kind of optimal PID controller has excellent control performance and robust performance.

关键词： fuzzy inference genetic algorithm fuzzy-GA PID controller incomplete derivation off-line on-line intelligent bionic artificial leg

Detecting direct associations in a network by information theoretic approaches

在线全文

学校读者我要写书评

暂无评论

Science China Mathematics 2019年第5期62卷 823-838页

作者： Jifan Shi Juan Zhao Tiejun Li Luonan Chen LMAM and School of Mathematical Sciences Peking University Laboratory of Systems Biology CAS Center for Excellence in Molecular Cell ScienceInstitute of Biochemistry and Cell Biology Shanghai Institutes for Biological SciencesUniversity of Chinese Academy of Sciences Chinese Academy of Sciences CAS Center for Excellence in Animal Evolution and Genetics Chinese Academy of Sciences

Detecting direct associations or inferring networks based on the observed data is an important issue in many fields, including biology, physics, engineering and social studies. In this work, we focus on the information theoretic approaches in the network reconstruction or the direct association detection, in particular,for biological networks. We not only review the traditional approaches or measurements on the associations among the observed variables, such as correlation coefficient, mutual information and conditional mutual information(CMI), but also summarize recently developed theories and methods. The new theoretic works include:information geometry to give a unified framework in detecting causality/association, the partial independence to alleviate the singularity of CMI, and multiscale analysis of CMI to avoid the underestimation issue of CMI.The new methods include part mutual information(PMI) and partial associations(PA), which improve the old measurements in avoiding both overestimation and underestimation. All those theories and methods make important contributions as major advances in the development of network inference.

关键词： network inference direct association information theory causal relation systems biology bioinformatics

Active Online Learning in the Binary Perceptron Problem

在线全文

学校读者我要写书评

暂无评论

Communications in Theoretical Physics 2019年第2期71卷 243-252页

作者：周海军 CAS Key Laboratory for Theoretical Physics Institute of Theoretical Physics Chinese Academy of Sciences

The binary perceptron is the simplest artificial neural network formed by N input units and one output unit, with the neural states and the synaptic weights all restricted to ±1 values. The task in the teacher-student scenario is to infer the hidden weight vector by training on a set of labeled patterns. Previous efforts on the passive learning mode have shown that learning from independent random patterns is quite inefficient. Here we consider the active online learning mode in which the student designs every new Ising training pattern. We demonstrate that it is mathematically possible to achieve perfect(error-free) inference using only N designed training patterns, but this is computationally unfeasible for large systems. We then investigate two Bayesian statistical designing protocols, which require 2.3N and 1.9N training patterns, respectively, to achieve error-free inference. If the training patterns are instead designed through deductive reasoning, perfect inference is achieved using N + log_2N samples. The performance gap between Bayesian and deductive designing strategies may be shortened in future work by taking into account the possibility of ergodicity breaking in the version space of the binary perceptron.

关键词： neural network perceptron online learning belief propagation statistical inference

Modelling on fuzzy control systems

在线全文

学校读者我要写书评

暂无评论

Science China Mathematics 2002年第12期45卷 1506-1517页

作者：李洪兴王加银苗志宏 Department of Mathematics Beijing Normal University Beijing 100875 China

A kind of modelling method for fuzzy control systems is first proposed here, which is calledmodelling method based on fuzzy inference (MMFI). It should be regarded as the third modelling method thatis different from two well-known modelling methods, that is, the first modelling method, mechanism modellingmethod (MMM), and the second modelling method, system identification modelling method (SlMM). Thismethod can, based on the interpolation mechanism on fuzzy logic system, transfer a group of fuzzy inferencerules describing a practice system into a kind of nonlinear differential equation with variable coefficients, calledHX equations, so that the mathematical model of the system can be obtained. This means that we solve thedifficult problem of how to get a model represented as differential equations on a complicated or fuzzy controlsystem.

关键词： fuzzy control systems, fuzzy inference rules, interpolation mechanism on fuzzy logicsystem, HX equations.

Interpreting forest diversity-productivity relationships:volume values,disturbance histories and alternative inferences

在线全文

学校读者我要写书评

暂无评论

Forest Ecosystems 2020年第1期7卷 64-75页

作者： Douglas Sheil Frans Bongers Faculty of Environmental Sciences and Natural Resource Management(MINA) Norwegian University of Life Sciences(NMBU)Box 50031432 AsNorway Forest Ecology and Forest Management Group Wageningen University&ResearchPO Box 476700 AA WageningenThe Netherlands

Understanding the relationship between stand-level tree diversity and productivity has the potential to inform the science and management of forests.History shows that plant diversity-productivity relationships are challenging to interpret—and this remains true for the study of forests using non-experimental field data.Here we highlight pitfalls regarding the analyses and interpretation of such studies.We examine three themes:1)the nature and measurement of ecological productivity and related values;2)the role of stand history and disturbance in explaining forest characteristics;and 3)the interpretation of any relationship.We show that volume production and true productivity are distinct,and neither is a demonstrated proxy for economic values.Many stand characteristics,including diversity,volume growth and productivity,vary intrinsically with succession and stand history.We should be characterising these relationships rather than ignoring or eliminating them.Failure to do so may lead to misleading conclusions.To illustrate,we examine the study which prompted our concerns—Liang et al.(Science 354:aaf8957,2016)—which developed a sophisticated global analysis to infer a worldwide positive effect of biodiversity(tree species richness)on“forest productivity”(stand level wood volume production).Existing data should be able to address many of our concerns.Critical evaluations will improve understanding.

关键词： Causation Correlation Diversity inference Productivity Richness Tree-growth Wood-density

Sample Size Affect Ethnobotanical Index Values: Bootstrap as a Remedial Approach

在线全文

学校读者我要写书评

暂无评论

Open Journal of Applied Sciences 2022年第11期12卷 1758-1769页

作者： Gbemavo Dossou Sèblodo Judes Charlemagne Cachon Fresnel Boris Lokonon Bruno Unité de Biostatistique et de Modélisation (UBM) Ecole Nationale Supérieure des Biosciences et Biotechnologies Appliquées (ENSBBA) Université Nationale des Sciences Technologies Ingénierie et Mathématiques (UNSTIM) Dassa-Zoumè Bénin Laboratoire de Biomathématiques et d’Estimations Forestières Faculté des Sciences Agronomiques Université d’Abomey-Calavi Cotonou Bénin Laboratoire de Biologie Expérimentale et Clinique (LaBEC) Ecole Nationale Supérieure des Biosciences et Biotechnologie Appliquées (ENSBBA) Université Nationale des Sciences Technologies Ingénierie et Mathématiques (UNSTIM) Dassa-Zoumè Bénin

Ethnobotanical indices are widely used to quantify cultural importance of plants in social studies. This study aims to show users of ethnobotanical indices the effect of sample variation and what methodological approach can be used to circumvent the problems related to sample variation. The methods used are to write an algorithm and used to simulate different sample sizes from which four ethnobotanic indices selected for the present study were estimated. Results showed the instability of the ethnobotanical indices under variations in the size of informants. It proposes bootstrapping as a statistical aid tool to remove the sample size effect in quantitative ethnobotany. For the indices used in the present study 1000 re-samplings eliminated the effect of sample size on the value of the indices. Researchers will have to take this new approach into account in order to calculate more precise ethnobotanical indices in order to better appreciate the cultural importance of plants.

关键词： Quantitative Ethnobotany Re-Sampling inference

The Novel and Hybrid Detection Approaches for V-BLAST Systems

维普期刊数据库评论

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

China Communications 2008年第4期5卷 143-149页

作者： Dong Wei Li Jiandong Chen Liang Xidian Univ Inst Informat Sci State Key Lab Integrated Serv Networks Broadband Wireless Communicat Lab Xian 710071 Peoples R China

Based on the new discrete particle swarm op-timization(NDPSO) and the mutation operation of the genetic algorithm,a hybrid discrete particle swarm op-timization(HDPSO) algorithm is proposed and applied to the signal detection of the Vertical Bell-Labs Layered Space-Time(V-BLAST)system.Then two detection ap-proaches that employ the HDPSO and parallel interfer-ence cancellation(PIC) algorithms in V-BLAST system are proposed.One of approaches is that the HDPSO algorithm is used as the first stage of the PIC to provide a good initial point for successive stages of the PIC,the other is that PIC is embedded into the HDPSO al-gorithm to improve further the fi tness of the population at each generation.Such a hybridization of the HDPSO with the PIC can speed up its convergence.In addition,a better initial data estimate supplied by the HDPSO algorithm improves the performance of the PIC,and the embedded PIC improves the performance of the HDP-SO.Analysis and simulation results show that the pro-posed detection approaches have lower computational complexity than the optimal ML detection approach and better detection performance than other conventional detection approaches.

关键词： Vertical Bell-Labs Layered Space-Time system Discrete particle swarm optimization Parallel inference cancellation